Abstarct: Today, many organizations are producing, storing, and transmitting a large amount of video data which, therefore, needs a large amount of memory; but, this is not commercially beneficial and also, takes large physical space. Therefore, it is difficult to displace and access such data storage devices.An obvious solution to this problem is to employ efficient video compression techniques. Region-Of-Interest (ROI) video coding is a newly introduced idea which has application in those video data in which all regions of a frxame have not equal importance. In this technique, more bit budget is spent for more important regions and thus, they will have better quality than other regions. It seems that the ROI video coding idea has acceptable performance in medical, sport, and personal images. In this thesis, a ROI soccer sport video compression method is proposed in which we define two different regions and encode them separately on the basis of their importance. The proposed method uses wavelet transform and encodes the wavelet coefficients using SPC coders especially SPIHT and EBCOT.Most important employed techniquesinclude classifying frxame into two types of independent and dependent ones using a shot boundary detection method, and encoding each type differently, using block matching technique in both spatial and transform domain, using block reordering technique, and finally, defining and determining interested regions in soccer video frxames. The compression efficiency of the proposed compression method is compared to that of MPEG4 and FLV video formats using PSNR, bit per pixel (bpp). And bit per second (bps) measures.Simulation results show that the proposed methods outperforms, in average, to the mentioned ones.